PERFORMANCE COMPARISON OF PYRAMIDAL HORNS LOADED … · 2018-01-14 · exhibits monotonically increasing gain behavior, a salient feature of the horn antenna [2,17,18], a return loss

Progress In Electromagnetics Research C, Vol. 17, 131–144, 2010

PERFORMANCE COMPARISON OF PYRAMIDAL HORNSLOADED WITH METAL BAFFLE OR WITH METAMA-TERIAL

C. Y. Tan and K. T. Selvan †

Electromagnetics and Telecommunication GroupThe University of Nottingham Malaysia CampusSemenyih 43500, Selangor, Malaysia

V. Venkatesan

SAMEER — Centre for ElectromagneticsCIT Campus, Taramani 600 113, Chennai, India

Abstract—Two recent methods that have been reported in theliterature to improve the performance of pyramidal horns are metalbaffle loading and the use of epsilon-near-zero metamaterial. In thispaper, a comparative study of the two methods is undertaken forthe case of Ku- and X-band horns. In addition to the simulationstudy, a C-band metal baffle loaded horn was fabricated and rigorouslycharacterized. It emerges from the study that E-plane metal baffleloading improves the radiation characteristics of the horn much betterthan the loading by metamaterial. Furthermore, the baffle loadingnearly retains the construction simplicity, weight and cost of thenormal pyramidal horn.

1. INTRODUCTION

The pyramidal horn is widely used as a standard gain horn, as a feedfor reflector and as an element in phased array antennas due to itssuch salient features as high gain, moderate bandwidth, constructionsimplicity, good power handling capability and low fabrication cost [1–4]. To make the pyramidal horn more compact, several approaches

Received 9 September 2009, Accepted 28 October 2010, Scheduled 8 November 2010Corresponding author: C. Y. Tan ([email protected]).

† C. Y. Tan is also with Amphenol Malaysia Sdn Bhd, Plot 1A, Bayan Lepas Free IndustrialZone 1, Penang 11900, Malaysia.

132 Tan, Selvan, and Venkatesan

that include lens-correction [5–7] and dielectric loading [8, 9] have beenconsidered. However, these compact horns, apart from being in generalheavy and expensive, present a high dielectric loss. In view of this,more recently, loading by metal baffle [10–12] and metamaterial [13–16] have been considered.

The idea of metal baffle loading to improve the performance ofthe pyramidal horn appears to have been first introduced in [10], andsubsequently investigated more thoroughly in [11]. It is shown in [11]that the performance of a short X-band pyramidal horn is substantiallyimproved when it is loaded with a pair of planar E- and H-plane metalbaffles. However the return loss of the horn thus loaded is high, and alsothe gain- versus-frequency plot exhibits multiple null regions. Thus theuseable bandwidth is reduced. In [12], triangular E-plane metal baffleloading approach has been proposed to overcome these shortcomingsof the planar metal baffles-loaded horn. The horn proposed in [12]exhibits monotonically increasing gain behavior, a salient feature ofthe horn antenna [2, 17, 18], a return loss of better than 15 dB over theentire X-band, and improved cross-polarization level.

According to [19], the first paper on metamaterial-loaded hornappeared in 1941. In that first paper (which, incidentally, has notbeen explicitly cited in [19]; however it is believed to be [20], availablein [1]), and also in [5], the gain of a short horn was shown to becomecomparable to that of optimum horn over a narrow bandwidth, when alayer of metal plates (metamaterial) was attached in front of the horn.The modern terminology for this type of metamaterial is epsilon-near-zero (ENZ) metamaterial or wire-medium [14]. After a lull, there hasnow been a revival of interest on the topic, with a number of articleshaving recently appeared on ENZ metamaterial-loaded pyramidal hornantennas [13–16]. In two of these efforts [13, 14], the metamaterial isinserted inside the normal pyramidal horn. While this approach doesshorten the horn, the bandwidth obtained is very narrow, matching isrelatively poor, and complexity is substantial.

In this paper, a simulation- and measurement-based study isundertaken on the performances of the triangular E-plane metalbaffle and ENZ metamaterial loaded normal pyramidal horns. Theresults show that the use of triangular E-plane metal baffle loadingoutperforms the metamaterial-loaded short horns with respect togain, aperture efficiency, radiation patterns, matching and bandwidth.At the same time, the resulting horn nearly preserves all thedesired features of normal pyramidal horn such as the normalgain behavior, bandwidth, simplicity and fabrication cost. It thusappears that the triangular E-plane metal baffle is an alternativeto the currently reported ENZ metamaterial loading techniques for

Progress In Electromagnetics Research C, Vol. 17, 2010 133

performance improvement of pyramidal horns. To validate theproposed technique, a metal baffle loaded C-band pyramidal horn isfabricated and rigorously measured.

The paper is organized as follows: Section 2 presents the geometryof the triangular E-plane metal baffle-loaded horn. The extensivesimulation results of the baffle-loaded horn and metamaterial-loadedpyramidal horns are compared in Section 3 while Section 4 comparesthe simulation and measurement results of a C-band unloaded andmetal baffle loaded horn. Section 5 discusses the results and theconclusion drawn in Section 6.

2. GEOMETRY OF THE TRIANGULAR E-PLANEMETAL BAFFLE LOADED PYRAMIDAL HORN

Figures 1(a) to (e) show the geometry and the physical outlook ofthe triangular E-plane metal baffle loaded pyramidal horn. Theperformance of the horn depends, apart from the horn dimensions, onthe baffle height (TBH), length (TBL) and position (LTB). All thesedimensions need to be optimized by using an appropriate simulationtool.

(a) (b)

(c) (d)


(g) (h)

(e) (f)

Figure 1. The geometry of the proposed triangular shaped E-planemetal baffle-loaded pyramidal horn: (a) Front view, (b) H-plane cross-section view, (c) E-plane cross-section view, (d) Photograph of anexperimental C-band baffle-loaded pyramidal horn, (e) Ku-band ENZmetamaterial lattice constant, (f) X-band ENZ metamaterial latticeconstant, (g) simulation model of the Ku-band ENZ loaded horn,(h) simulation model of the X-band ENZ loaded horn.

An experimental C-band baffle-loaded horn is made by insertinga baffle into a C-band short pyramidal horn available in the lab (thishorn had been used in [21]). Parametric studies were carried out on theaxial location of the triangular E-plane metal baffle (LTB) for chosenbaffle height (TBH) and given length (TBL) of the short horns. Thedimensions are given in Table 1.

The two epsilon-near-zero metamaterial-loaded short pyramidalhorns considered in [13, 14] are chosen for the present simulation-based comparative study. For ready reference, the metamaterial latticeconstants used in [13] and [14] are given in Fig. 1(e) and Fig. 1(f),respectively; more details can be found in these references. In boththe cases, the metamaterial is placed inside the horn aperture; seeFig. 1(g) and Fig. 1(f). While the Ku-band ENZ metamaterial has aplasma frequency of 15.81 GHz with the corresponding epsilon valuesof 0 to 0.23 (from 15.81GHz to 18 GHz), the X-band ENZ has a plasma


frequency of 9.3 GHz with the corresponding epsilon values of 0 to 0.44(from 9.3 GHz to 12.4GHz).

Table 1. Dimensions of the triangular E-plane metal baffle-loadedhorns.

LabelDimension in (mm)

A B a b L LTB TBH TBL T

Baffle Loaded

Horn 1

(Ku-band)

75.49 67.59 15.79 7.90 64.00 12.00 13.00 13.00 2.00

Baffle Loaded

Horn 2

(X-band)

125.00 112.00 22.50 10.00 99.00 15.00 19.50 19.50 1.00

Baffle Loaded

Horn 3

(C-band)

288.00 213.50 35.00 16.00 259.00 30.00 40.71 40.71 3.00

Figure 2. Simulated gains ofthe Ku-band Horns: Optimumgain horn, 44% Shorter Horn(short horn 1), metamaterialloaded short horn (Metamaterialloaded horn 1) and the proposedtriangular E-plane metal baffle-loaded short horn (Baffle loadedhorn 1).

Figure 3. Simulated gains ofthe X-band Horns: Optimumgain horn, 48% Shorter Horn(short horn 2), metamaterialloaded short horn (Metamaterialloaded horn 2) and the proposedtriangular E-plane metal baffle-loaded short horn (Baffle loadedhorn 2).


3. SIMULATED RESULTS

Extensive simulation studies by using the commercial 3D electromag-netic simulation package CST Microwave Studio were undertaken tocompare the performance of the triangular E-plane metal baffle-loadedhorns with that of pyramidal horns loaded with metamaterial.

Figures 2 and 3 displayed the simulated gains in the Ku- andX-band of unloaded horns and horns loaded with triangular E-planemetal baffle and ENZ metamaterial. As could be seen, the baffle-loaded horns offer a gain improvement of about 3 dB through the bandswhile retaining the normal gain behavior of the optimum gain horns.On the other hand, the metamaterial-loaded horns offer similar gainimprovement only over several narrow bands.

The aperture efficiency of the horns is compared in Figs. 4 and5 for the Ku- and X-band horns respectively. Understandably, it isobserved that the efficiency of the triangular E-plane metal baffle-loaded horns is better than the unloaded and metamaterial loadedhorns throughout the full bands. Besides, the baffle-loaded horns offercomparable efficiencies as the unloaded optimum gain pyramidal hornat much shorter length.

As should be expected, the return loss of both the metamaterial-and metal baffle-loaded horns would be higher than that of theunloaded horns. This is seen in Figs. 6 and 7 for the Ku- and X-band horns respectively. However, the baffle-loaded horns performmuch better than the metamaterial-loaded horns in respect of returnloss too. As has been observed through simulation, the return loss ofthe baffle-loaded horn can be improved to better than 15 dB by furtheroptimizing the tapered section of the baffle, but at the price of slightlylower gain.

Figure 4. Aperture efficienciesof the Ku-band horns.

Figure 5. Aperture efficienciesof the X-band horns.


Figure 6. Return loss of the Ku-band horns.

Figure 7. Return loss of the X-band horns.

The E- and H-plane radiation patterns of the different hornsconsidered in the study are shown for the Ku- and X-band in Figs. 8 and9 respectively. The patterns are shown at the band-edge, mid-band andresonant frequencies. As can be generally seen, the baffle-loaded hornsoffer much better E-plane patterns (with respect to directionality andsidelobes level) than the other horns. The H-plane patterns of baffle-loaded horns are broader than the optimum horns but narrower thanthe unloaded short horns.

4. EXPERIMENTAL SETUP AND MEASUREMENTRESULTS

The modified three antenna method [21] is applied to characterizethe gain of the C-band unloaded and baffle-loaded short pyramidalhorn. The measurement was conducted in SAMEER-Center forElectromagnetics, Chennai. The three horns comprised a commercialC-band standard gain horn as the reference antenna and two pyramidalhorns, one of them with metal baffle-loading. The separation distancebetween the transmitting and receiving horns was 5m and the hornswere mounted at 2.5 m height. With this setup, power measurementwas repeated ten times for the different combination of the horns.The 3σ measurement uncertainty level in the frequency band of 5.8to 8.2 GHz is believed to be within ±0.5 dB.

Figure 10 shows the measured and simulated gain of the unloaded-and proposed metal baffle-loaded horns. As seen from the figure, thegain of the pyramidal horn with metal baffled-loading was higher by0.8 dB to 5.2 dB over the entire C-band frequencies. Also, the simulatedand measured results are in good correlation. The respective apertureefficiency of the two horns is show in Fig. 11. As expected, the apertureefficiency of the proposed metal baffle-loaded horn is higher comparedto that of the unloaded horn.


E-plane H-plane1

2.4

GH

z 1

5.2

GH

z

18

GH

z 1

6.4

GH

z

Figure 8. The principal E- and H-plane radiation patterns of thethree horns at the band edges and center frequency of the Ku-band.


E-plane H-plane8

.2 G

Hz

9.7

GH

z 1

0.3

GH

z 1

2.4

GH

z

Figure 9. The principal E- and H-plane radiation patterns of thethree horns at the band edges and center frequency of the X-band.


The return loss of the two horns is shown in Fig. 12. The mismatchbetween the measured and simulated results is believed to be due tothe coaxial adaptor not being included in the simulation model. Whileconsidering the measured return loss of the unloaded (short horn 3)and metal baffle loaded horns, it is observed that the baffle loadingmarginally worsens the return loss of the unloaded horn. This findingis consistent with Figs. 6 and 7.

Figure 13 shows the measured and simulated E- and H-plane ofthe two horns. As observed from Fig. 13, baffle loading narrows the E-plane beamwidth and lowers the side-lobes level of the unloaded horn.Interestingly, the baffle loading is also seen to be useful in eliminatingthe E- and H-plane pattern squint of the unloaded horn at 8.2 GHz.

Figure 10. Measured andsimulated gains of the C-bandhorns.

Figure 11. Measured andsimulated aperture efficiency ofthe C-band horns.

Figure 12. Measured and simulated return loss of the C-band horns.


E-plane H-plane5

.8 G

Hz

7.0

GH

z 8

.2 G

Hz

Figure 13. Measured and simulated E- and H-plane radiation patternof the unloaded and baffle-loaded horns.

5. DISCUSSION

In an attempt to better understand the reasons for the aboveobservations, Figure 14 shows the simulated electric field distributionsin the E-plane (Y Z-plane) and the H-plane (XZ-plane) at 10.3 GHzfor the unloaded short horn and also for the horns loaded withtriangular E-plane metal baffle and metamaterial. As seen, the E-plane wave fronts of the baffle-loaded horn are more uniform thanthose of the other horns. However, the same cannot be said of the H-


E-plane H-planeS

hort

Horn

M

etam

ater

ial-

load

ed h

orn

B

affl

e-lo

aded

H

orn

Figure 14. Electrical field distributions of the three horns at thecenter frequency (10.3 GHz) of X-band.

plane wave fronts. It may be deduced that loading by H-plane bafflewill possibly result in more uniform wave fronts in that plane. Animplication is that the E- and H-plane phase errors of the short horncan be corrected by using appropriate metal baffle loadings.

6. CONCLUSION

In this paper, a simulation-based comparative study was undertakenon the performance of X- and Ku-band pyramidal horns loaded withtriangular E-plane metal baffle or with epsilon-near-zero metamaterial.In general, in both the bands (X- and Ku-band), the baffle-loadedhorn was seen to offer better performance than the metamaterial-loaded horn. In addition to the simulation study, an E-plane metalbaffle-loaded horn operating in C-band was fabricated and rigorouslycharacterized to validate the simulation findings.

While the study does demonstrate that the proposed triangular


E-plane metal baffle loading technique is more effective to improvethe performance of a short pyramidal horn compared to with ENZmetamaterial-loading, is it by no means to discourage the research onmetamaterial, indeed it indirectly encourage to research on a bettertype of metamaterial to improve the horn performance. Surprisingly,the findings could be in line with [22] where the authors showedthat the dielectric- and Frequency Selective Surface (FSS)-superstratesloading methods are more effective to improve the microstrip antennadirectivity compared to double-negative (DNG) metamaterial. Also,the directivity improvement bandwidth of dielectric-superstrate isgreater than FSS-superstrate microstrip antenna.

In short, from a practical point of view, the baffle-loadingentails much lower fabrication complexity than the considered ENZmetamaterial loading, with corresponding cost implication. Moreover,the proposed technique offers better radiation characteristics comparedto the considered metamaterial-loading. The proposed horn may beuseful as a feed element in phased array system.

REFERENCES

1. Love, A. W., Electromagnetic Horn Antennas, IEEE Press, NewYork, 1976.

2. Stutzman, W. L. and G. A. Thiele, Antenna Theory and Design,2nd edition, John Wiley & Sons, 1998.

3. Balanis, C. A., Antenna Theory Analysis and Design, 3rd edition,John Wiley & Sons, New Jersey, 2005.

4. Bird, T. S. and A. W. Love, “Horn antennas,” AntennaEngineering Handbook, 4th edition, 14-1–14-74, J. Volakis (ed.),McGraw-Hill, 2007.

5. Kock, W. E., “Metal-lens antennas,” Proceedings of the IRE,Vol. 34, 828–836, 1946.

6. Chatterjee, R., Antenna Theory and Practice, 2nd edition, NewAge International, 1996.

7. Elliott, R. S., Antenna Theory and Design, Revised edition, JohnWiley & Sons, 2003.

8. Lier, E. and R. Landes, “High gain linearly polarised dual bandhorn for multi-feed reflector antenna,” Electronics Letters, Vol. 39,1230–1232, 2003.

9. Tsandoulas, G. and W. Fitzgerald, “Aperture efficiency enhance-ment in dielectrically loaded horns,” IEEE Transactions on An-tennas and Propagation, Vol. 20, 69–74, 1972.

10. Risser, J. R., “Waveguide and horn feeds,” Microwave Antenna


Theory and Design, 1st edition, S. Silver (ed.), 383, McGraw-Hill,1949.

11. Koerner, M. A. and R. L. Rogers, “Gain enhancement of apyramidal horn using E- and H-plane metal baffles,” IEEETransactions on Antennas and Propagation, Vol. 48, 529–538,2000.

12. Tan, C. Y. and K. T. Selvan, “Performance improvementof pyramidal horn using triangular E-plane metal baffle,”3rd European Conference on Antennas and Propagation, 2009.EuCAP 2009, 3034–3037, 2009.

13. Wu, Q., P. Pan, F.-Y. Meng, L.-W. Li, and J. Wu, “Anovel flat lens horn antenna designed based on zero refractionprinciple of metamaterials,” Applied Physics A: Materials Science& Processing, Vol. 87, No. 2, 151–156, May 2007.

14. Hrabar, S., D. Bonefacic, and D. Muha, “ENZ-based shortenedhorn antenna — An experimental study,” IEEE Antennas andPropagation Society International Symposium, 2008. AP-S 2008,1–4, 2008.

15. Hwang, R.-B., H.-W. Liu, and C.-Y. Chin, “A metamaterial-basedE-plane horn antenna,” Progress In Electromagnetics Research,Vol. 93, 275–289, 2009.

16. Xiao, Z. and H. Xu, “Low refractive metamaterials for gainenhancement of horn antenna,” Journal of Infrared, Millimeterand Terahertz Waves, Vol. 30, 225–232, March 2009.

17. Selvan, K. T., “Derivation of a condition for the normal gainbehavior of pyramidal horns,” IEEE Transactions on Antennasand Propagation, Vol. 48, 1782–1784, 2000.

18. Tan, C. Y., K. T. Selvan, and Y. M. Koh, “Derivation of conditionsfor the normal gain behavior of conical horns,” InternationalJournal of Antennas and Propagation, Vol. 2007, 4, 2007.

19. Ramsay, J., “Highlights of antenna history,” IEEE Antennas andPropagation Society Newsletter, Vol. 23, 7–20, 1981.

20. Rust, N. M., “The phase correction of horn radiators,” J. Inst.Elec. Eng., Vol. 93, 50–51, Mar–May 1946.

21. Selvan, K. T., “A modified three-antenna gain measurementmethod to simplify uncertainty estimation,” Progress In Electro-magnetics Research, Vol. 57, 197–208, 2006.

22. Mittra, R., Y. Li, and K. Yoo, “A comparative study of directivityenhancement of microstrip patch antennas with using threedifferent superstrates,” Microwave and Optical Technology Letters,Vol. 52, No. 2, 327–331, February 2010.

Documents

PERFORMANCE COMPARISON OF PYRAMIDAL HORNS LOADED … · 2018-01-14 · exhibits monotonically increasing gain behavior, a salient feature of the horn antenna [2,17,18], a return loss